The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute gh (GitHub CLI) and git commands. These commands are employed to query repository metadata, fetch issue comments, and identify modified files for security review checks. This behavior is consistent with the skill's stated purpose of enforcing a pull request gate.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources by retrieving comments from GitHub issues via the API. It parses this content using grep and jq to look for specific markers () and compliance status. While this ingestion represents a surface for indirect instructions, the implementation uses restricted parsing logic rather than direct prompt interpolation or code execution of the external content, minimizing potential risk.

review-gate